EUVIMEDEUVIMED entdecken
Multizentrische StudieOpen Access · CC BY 4.0

Image Quality Assessment of Diffusion-Weighted Imaging (DWI) and Its Impact on Apparent Diffusion Coefficient (ADC) as a Quantitative Imaging Biomarker for Predicting Response to Neoadjuvant Chemotherapy in High-Risk Early Breast Cancer

Wen Li iD, Lisa J Wilmes, Julia Carmona-Bozo, Nu N Le, Maggie Chung, Jessica E Gibbs iD, Natsuko Onishi iD, Elissa Price, Bonnie N Joe iD, John Kornak iD, Thomas L Chenevert, Dariya Malyarenko iD und 3 weitere

Tomography · 17.06.2026 · 12 (6) · 87

DOI: 10.3390/tomography12060087 ↗

Kurz gefasst

Multizentrische Studie in der Zeitschrift Tomography, online erschienen am 17.06.2026. Verfasst von Wen Li und 14 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiology and Biomedical Imaging, University of California, San Francisco, CA 94158, USA. Hauptthemen laut MEDLINE: Diffusion Magnetic Resonance Imaging, Breast Neoplasms, Neoadjuvant Therapy. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 32 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NCI NIH HHS, NIH/NCI, NIH/NIBIB.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background/Objectives: Apparent diffusion coefficient (ADC) calculated from diffusion-weighted MRI (DWI) can predict tumor response to neoadjuvant chemotherapy for breast cancer. However, obtaining consistently adequate image quality in breast DWI can be challenging, and the effect of image quality on ADC’s predictive performance is unclear. The objective of this study was to evaluate inter-reader variability in image quality assessment and the effect of DWI image quality on the predictive performance of ADC.

Methods

This multi-institutional study included 428 patients. Two readers assessed three DWI image quality factors—fat suppression, artifacts, and signal-to-noise ratio (SNR). Inter-reader agreement was estimated using Fleiss’ Kappa. The percent change in tumor ADC from pretreatment (T0) to early treatment (T1) was used to predict pathologic complete response (pCR), assessed at surgery.

Results

Out of 428 patients, 134 were excluded (missing pCR [n = 17]; missing/incorrect DWI [n = 23]; inability to define region-of-interest [ROI, n = 94]) and 294 were included in the analysis. Kappa coefficients were estimated as: 0.47 (95% confidence interval [CI]: 0.42, 0.52) for fat suppression, 0.54 (0.50, 0.59) for artifact, and 0.38 (0.32, 0.44) for SNR. The AUC of ADC calculated from DWI with adequate (high or medium at both time points) image quality was 0.61 (95% CI: 0.52, 0.702), while it was 0.68 (95% CI: 0.53, 0.83) from DWI with inadequate image quality at either T0 or T1. The p-value for the difference in AUCs was 0.45.

Conclusions

The inter-reader agreement was moderate to fair across all three quality categories. When a manually delineated tumor ROI was possible, no statistically significant difference in ADC predictive performance was observed between the quality-adequate and quality-inadequate cohorts; still, both were predictive of pCR. Furthermore, no statistically significant differences were observed in inter-reader agreement or ADC predictive performance between 1.5T and 3T scanners. These findings are clinically relevant to the use of ADC as an imaging biomarker in real-world conditions.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Diffusion Magnetic Resonance ImagingBreast NeoplasmsNeoadjuvant TherapyHumansFemaleMiddle AgedPredictive Value of TestsPathologic Complete ResponseSignal-To-Noise RatioAdultAgedObserver VariationArtifactsChemotherapy, AdjuvantTreatment OutcomeRetrospective Studies

Stichwörter der Autor:innen

apparent diffusion coefficientartifactbreast cancerdiffusion-weighted MRIfat suppressionimage qualitypathologic complete responsesignal-to-noise ratio

Autor:innen und Institutionen

  • Department of Radiology and Biomedical Imaging, University of California, San Francisco, CA 94158, USA.
  • Department of Epidemiology and Biostatistics, University of California, San Francisco, CA 94158, USA.
  • Department of Radiology, University of Michigan, Ann Arbor, MI 48109, USA.
  • Department of Radiology, University of Minnesota, Minneapolis, MN 55455, USA.
  • Department of Radiology, University of Washington, Seattle, WA 98195, USA.

ORCID-Profile: Wen Li, Jessica E Gibbs, Natsuko Onishi, Bonnie N Joe, John Kornak, Dariya Malyarenko, Patrick J Bolan, Savannah C Partridge

Förderung

  • NCI NIH HHS
  • NIH/NCI
  • NIH/NIBIB

Zitieren

APA

Li, W., Wilmes, L. J., Carmona-Bozo, J., Le, N. N., Chung, M., Gibbs, J. E., Onishi, N., Price, E., Joe, B. N., Kornak, J., Chenevert, T. L., Malyarenko, D., Bolan, P. J., Partridge, S. C., & Hylton, N. M. (2026). Image Quality Assessment of Diffusion-Weighted Imaging (DWI) and Its Impact on Apparent Diffusion Coefficient (ADC) as a Quantitative Imaging Biomarker for Predicting Response to Neoadjuvant Chemotherapy in High-Risk Early Breast Cancer. Tomography, 12(6), 87. https://doi.org/10.3390/tomography12060087

Vancouver

Li W, Wilmes LJ, Carmona-Bozo J, Le NN, Chung M, Gibbs JE, et al. Image Quality Assessment of Diffusion-Weighted Imaging (DWI) and Its Impact on Apparent Diffusion Coefficient (ADC) as a Quantitative Imaging Biomarker for Predicting Response to Neoadjuvant Chemotherapy in High-Risk Early Breast Cancer. Tomography. 2026;12(6):87. doi:10.3390/tomography12060087

BibTeX
@article{Li2026,
  author = {Li, Wen and Wilmes, Lisa J and Carmona-Bozo, Julia and Le, Nu N and Chung, Maggie and Gibbs, Jessica E and Onishi, Natsuko and Price, Elissa and Joe, Bonnie N and Kornak, John and Chenevert, Thomas L and Malyarenko, Dariya and Bolan, Patrick J and Partridge, Savannah C and Hylton, Nola M},
  title = {Image Quality Assessment of Diffusion-Weighted Imaging (DWI) and Its Impact on Apparent Diffusion Coefficient (ADC) as a Quantitative Imaging Biomarker for Predicting Response to Neoadjuvant Chemotherapy in High-Risk Early Breast Cancer},
  journal = {Tomography},
  year = {2026},
  volume = {12},
  number = {6},
  pages = {87},
  doi = {10.3390/tomography12060087}
}
Kostenlos starten →